itiona1 The causes and effects of aging in quick clays

.57.
:r Hat
Land- GHISLAIN
LESSARD
GPologie et MPcanique des Sols, SociCtP d ' ~ n e r ~de
i e la Baie James, 800 est, boulevard de Maisonneuve, MontrPal, QuP.,
%rice Canada H2L 4M8
emoir AND
JAMES
K . MITCHELL
Fraser
Department of Civil Engineering, University of California, Berkeley, CA 94720, U.S.A.
~gical
Received November 19, 1984
'logy, Accepted April 3, 1985
Paper
The physicochemical properties of a clay from La Baie, Quebec were found to be typical of Champlain quick clays: very low
)f the plasticity, liquidity index greater than 3, sensitivity greater than 500, and pH of about 9.5. The mineralogy is characterized by the
oject, abundance of primary or "rock flour" minerals such as feldspar, quartz, amphibole, and calcite, with illite as the principal clay
mineral. The pore water chemistry consists predominantly of sodium bicarbonate. Low concentrations of sulfate, calcium, and
Ytton, magnesium and high alkalinity and pH are attributed, in part, to anaerobic sulfate-reduction processes taking place in situ.
?. Quick clay samples stored in the laboratory showed signs of aging regardless of the storage procedure. The remoulded strength
mass- and the liquid limit increased with time, whereas the sensitivity, the liquidity index, and the pH decreased. The water content,
views plastic limit, and undisturbed strength remained practically unchanged. The pore water concentrations of calcium, magnesium,
and sulfate increased by severalfold.
or the The aging phenomenon is attributed, for the most part, to the oxidation of iron sulfide, which results in the formation of iron
:oard, hydroxide and sulfuric acid. The production of acid causes the dissolution of calcium carbonate, which increases the
concentration of divalent cation in the clay, thereby reducing interparticle repulsion and increasing the remoulded strength. The
istory oxidation of organic matter, resulting in the formation of carbonic acid, also contributes to the aging by its effect on the solubility
~lletin of calcium carbonate.
Key words: aging, chemical reactions, physicochemical phenomena, sensitivity, time effects.
ley of
I1 a t t t constatt que les proprittts physico-chimiques d'une argile de Baie, QuBbec Btaient typiques des argiles sensibles
llooet Champlain: trks faible plasticitt, indice de liquiditt sup6rieure ii 3, sensibilitt sup6rieure & 500 et pH d'environ 9,5. La
s and mintralogie est caracttriste par l'abondance de "poussikre de roche" ou de mintraux primaires tels que le feldspath, le
quartz, les amphiboles et la calcite. L'illite reprtsente le principal mintral argileux. La chimie de l'eau interstitielle est constitute
mica1 principalement de bicarbonate de sodium. Les faibles concentrations de sulfate, de calcium et de magnesium, de m&meque la
urvey forte alcalinitt et la valeur t l e v k du pH sont attributes, en partie, aux processus de rBduction anatrobique du sulfate qui se
" produit in situ.
,ma- Des tchantillons d'argile sensible entreposts en laboratoire ont montrt des signes de vieillisement indtpendarnment de la
proctdure d'entreposage. La rtsistance remanite et la limite liquid ont augmentt avec le temps alors que la sensibilitt, l'indice de
's and liquiditt et le pH diminuaient. La teneur en eau, la limite plastique et la rtsistance intacte sont demeukes ii peu prks
1 PP. inchangtes. Les concentrations de calcium, de magntsium et de sulfate dans l'eau interstitielle ont augmentt drastiquement.
Le phtnomtne de vieillissement est attribut, en grande partie, 21 I'oxydation du sulfure de fer qui rtsulte en la formation
ntrol. d'hydroxide de fer et d'acide sulfurique. La production d'acide cause la dissolution du carbonate de calcium, et augmente la
concentration de cations bivalents dans I'argile, rtduisant ainsi la rtpulsion inter-particule et augmentant la rtsistance remanite.
L'oxydation de matitre organique, rtsultant en la formation d'acide carbonique, contribue aussi au vieillissement par suite de
son effet sur la solubilitk du carbonate de calcium.
Mots ~14s:vieillissement, kactions chimiques, phtnomknes physico-chimiques, sensibilitt, effet du temps.
Can.h t e c h . J. 22,335-346 (1985)

Introduction
Soil engineering studies often include laboratory tests aimed
at predicting the behaviour of the in situ soil. Generally,
undisturbed samples are taken from the site and transported to
the laboratory, where they are stored at room temperature and
100% relative humidity to prevent water losses. Whether the
properties measured by laboratory tests carried out weeks to
months after sampling are representativeof the in situ character-
istics of the soil is questionable. Many cases have been reported
where significant changes in sensitivity, Atterberg limits, and
pore water chemistry occurred during storage. The process
responsible for these changes, generally referred to as aging,
has been observed in Scandinavian as well as Canadian clays.
Bjerrum and Rosenqvist (1956) observed an increase in the
Atterberg limits of a marine clay from Asrum, Norway, that was
a result of aging. The plasticity index of a sample rose from 21%
to 27% over a 2 year period and was accompanied by an increase
in potassium ion concentration in the pore water, which was
attributed to the weathering of the clay minerals.
Samples of Swedish clays taken from the Gota River valley
lost their original quickness after being stored for 2-4 years.
Similar sensitivity reduction was observed in situ, as a clay
sampled in 1967 at a depth of 1 m below an exposed slide bottom
of the 1960 slide at Veston had lost its quickness in the
meantime (Siiderblom 1969). Quantitative analyses by standard
paper chromatography showed that before aging, the quick clay
contained mainly Na+ and Sod2- in its pore water. After aging,
there was a significant amount of ~ g and~Ca2+ + and a marked
increase of S042- (Sijderblom 1974). According to Siiderblom
(1969), "the change from anaerobic to aerobic state and the
accompanying changes in the microbial activity may be the
most important factor in the aging process."
Torrance (1976) reported the effects, on the pore water
chemistry, of 3 months' storage under a wide variety of standard
and modified storage procedures. The material was a soft
Champlain clay with a low salinity and a sensitivity of 10-20.
Irrespective of the storage method, the salinity and the percent-
age of divalent cations in the pore water had increased. The most
 CAN. GEOTECH.J. VOL. 22, 1985

STORAGE PROCEOURE
A: Shelby tube, ends waxed ( 4 samples
from tube)
B: Aluminum f o l d , waxed ( d u p l i c a t e
sanples)
C: P l a s t i c wrap waxed ( d u p l i c a t e
samples)
D: Yaxed
E: Container N2
F: Container, p l a s t i c wrap N,
G: Container, a i r I I I

ORlGlllAL AVERAGE A B C 0 E F G
STORED

FIG.1. Effect of 3 months' storage upon the cation distribution in the pore water under various storageprocedures (adapted from Torrance 1976).

conspicuous changes were observed for the clay stored in plastic Experimental methods and material characteristics
containers without the protective seal of the wax (Fig. 1). Upon reception of the samples in the laboratory, about 1
According to Torrance, the increase in the concentration of month after sampling, an extensive analysis of the geotechnical
calcium and magnesium in the pore water during aging is and physicochemical properties of the quick clay was under- Most
probably related to attack on carbonates present in the soil. taken. The purposes of the analysis were firstly to evaluate increa
In order to better understand the geotechnical and chemical whether any change had occurred in the geotechnical properties ' montl
changes occurring in a quick clay during storage, an extensive of the material since sampling and secondly to establish the in The
testing program was carried out on a Champlain quick clay situ or "initial" geochemical characteristics of the clay, which ' from ;
from La Baie, Quebec. After measuring the initial charac- were to be used as a reference to assess the various effects of thoro~
teristics of the material, samples were stored and tested aging. The te
periodically over a period of 1 year to evaluate the effect of indica
various storage conditions. The main results of the testing Material description
The soil material is an extremely quick, banded silty clay of mater!
program are presented in this paper. Their interpretation and so liq~
significance in geotechnical practice are emphasized. medium to stiff consistency. Its colour varies from light to dark
bluish gray with a few brownish bands. Most of the bands are was g
Sampling horizontal but some are inclined by as much as 45" from the streng
The quick clay samples used in the testing program were horizontal. Their thicknesses range from less than 1 cm to about SpecQ
obtained from La Baie near the Saguenay River in the Province 30 cm. Pockets and lenses of fine to medium sand are encoun- The
of Quebec. The clay now covering most of the Saguenay - Lac tered randomly throughout the strata. It is important to note that varied
Saint-Jean area was desposited in the LaFlamme Sea, an the bands do not necessarily reflect large differences in grain miner.
extension of the Champlain Sea, after the last glaciation some size and mineralogy, as is the case for freshwater varved clays, The
10 000 years ago. The sampling site is located about 20 km from which usually show an alternation of horizontal layers of silt and are s h ~
Saint-Jean-Vianney where the largest landslide known to have clay. In the case of the La Baie clay, the darker bands are 40%
occurred in the Champlain clay took place some 400-500 years probably related to the presence of metastable iron sulfide. surpri
ago and involved 20km2 of land, displacing an estimated Water content and Atterberg limits water
volume of 200 X lo6 m3 of soil. In 1971, a flow slide occurred The distribution of water content with depth (Fig. 2) is typical Orgat
in the same crater, causing the loss of 31 lives and canying 40 of the heterogenity of Champlain clays. Variation in water The
houses to destruction (Tavenas et al. 1971). content of as much as 10% may occur within a distance of a few titratic
The samples were recovered in October 1979 using 76 mm centimetres, but the general trend shows little variation with carbo1
diameter Shelby tubes. Within 1 week after sampling, the clay depth. Most values fall between 40% and 50%. addinl
was extruded from the sampling tube. The sample surface that The Atterberg limits measured 1 month after sampling did not potass
had been in contact with thesteel tube was removed by scraping,
in order to avoid any potential contamination by metallic ions.
differ significantly from those measured within 1 week of
sampling. The plastic limit varies between 16% and 23%, with
I1 dichrc
detem
The clay was then cut horizontally in 10 cm long samples, which an average value of 19% and the liquid limit ranges from 20% to The
,
I
were waxed with several thin layers of paraffin. In order to 33%, with an average of 25%. The plasticity index is very low organ;
monitor any change in properties that occurred during transport with values ranging from 4% to lo%, averaging 6%. The averal
from La Baie, Quebec to Berkeley, California, some samples liquidity index varies from 3 to 7 and reflects the extreme I carbol
were tested immediately after the extrusion of the clay from the quickness of the clay. I

tube. Attenberg limits were determined following standard Carbc

procedures and without predrying the soil. Undisturbed and Stength and sensitivity I
Thc
remoulded strengths were measured by means of the Swedish The undisturbed strength, determined by means of the the ral
fall-cone test. fall-cone test, represents the average of five measurements.
 LESSARD AND MITCHELL

Geotechnical and compositional characteristics of La Baie clay measured 1 week (A) and 1 month (0)
after sampling.
,out 1
mica1
inder- Most values fall between 50 and 70kPa with a very slight of acid that is left after reaction with the carbonate is determined
iluate increase with depth. The undisturbed strength measured at 1 by NaOH titration.
crties month is essentially the same as at 1 week after sampling. The tests indicated carbonate contents varying between 7%
the in The remoulded strength was determined by the fall-cone test and 9%, with an average value of 8%. No particular trend with
which from at least five measurements. Before each test, the soil was depth was observed. Such high values of carbonate content are
cts of thoroughly remoulded in order to minimize thixotropic effects. common for the LaFlamme Sea clays.
The test was repeated until a constant reading was obtained, thus
indicating a complete remoulding and homogenization of the Soil pH
material. For all the tests performed, the clay samples became The soil pH was measured in a 1: 1 (by weight) mixture of soil
lay of and distilled water. Five grams (dry weight) of wet soil of
1 dark so liquid upon remoulding that the penetration of the 10 g cone
was greater than 20 mm, which corresponds to a remoulded known water content was used for the test. Distilled water was
is are added to the soil to obtain a total weight of 10 g. The slurry was
n the strength lower than 0.07 kPa and a sensitivity greater than 500.
mixed thoroughly and then allowed to stand undisturbed for 1 h,
about Specijic gravity and grain-size distribution after which the pH was measured by means of a combination
coun- The specific gravity of the soil solids measured on six samples electrode and a Beckrnan pH-meter positioned in near-contact
e that varied only from 2.75 to 2.78, which reflects a homogeneous with the bottom of the container. Typically, the reading
grain mineralogical composition of the material. decreased rapidly to reach a stable value within 2-3 min.
:lays, The results of grain-size analyses performed on six samples The soil pH measured in the clay samples decreases with
It and are shown in Fig. 2. The clay-size fraction (-2 ~ mvaries
) from depth, with values ranging from 9.6 in the upper part of the
is are 40% to 6296, with the smaller values at mid-depth. Not boring to 9.3 in the lower part.
". surprisingly, the grain-size distribution correlates well with the
water content of the material and plastic limit profile. Mineralogy
The mineralogical analysis of the clay was performed by the
.pica1 Organic matter
X-ray diffraction method using a RIGAKU diffractometer and
water The organic matter content of the clay was measured by redox Cu K, radiation. Six samples from various depths were air-dried
a few titration (Rankin 1970). The method is based on the oxidation of and reduced to powder. The material passing the #200 mesh
with carbon by potassium dichromate. The procedure consists of sieve was packed into the sample holder and the surface struck
adding concentrated sulfuric acid and a known amount of
off smooth and level. The X-ray patterns obtained for the six
,dnot potassium dichromate to 5.0g of dry soil. The amount of
powder samples indicate that the mineralogical composition of
:k of dichromate that does not react with the oxidizable carbon is the material is virtually the same throughout the profile. As
with determined by titration.
shown in Fig. 3a, the material contains a substantial amount of
)% to The results showed no particular trend with depth. The rock flour minerals such as quartz, feldspar, amphibole, and
/ low organic matter content varied between 0.4% and 0.6%, with an calcite. Illite is the main phyllosilicate mineral. Chlorite and
The average of 0.5%, which corresponds to 0.3% of oxidizable kaolinite are also present in traces.
reme Carbon.
A semiquantitative X-ray analysis was performed on one
Carbonate sample. Approximately 20g of air-dried soil was ground for
The total insoluble carbonate content was determined using 2 min in a rotative grinder. The powder obtained was packed in
the the rapid titration method (Hesse 1971). The procedure consists the sample holder, smoothed off, and pressed with fine-grained
ents. of adding 50 mL of 1 N HC1 to 5 g of air-dried soil. The amount sandpaper to minimize the preferential orientation of platy mica
 338 CAN. GEOTECH. 1. VOL. 22. 1985
I
Cu Ko RADIATION. = 1.54 H
Q: quartz
F: feldspar
A: amphibole
C: calcite
I: illite
K: kaolinite
Ch: chlorite

l , , , , l , , , , ~ , , , , ~ , , , , ~ , , , , ~ , , , , ~
5 10 15 20 25 28 30 35
(a1 Silt- and clay-size fraction ( 4 2 0 0 ) 1I

(b) Clay-size fraction ( <2um )

FIG.3. X-ray diffraction powder patterns of fraction passing #200 sieve ( a ) and oriented <2 km fraction (b) of sample from La Baie.
Th
and amphiboles (Foscal-Mella 1976). Reference standard TABLE1. Mineralogical composition leach
minerals were prepared following the same procedure as for the of La Baie quick clay nece:
clay. Biotite was used to estimate the illite content, since the two the ci
have similar structures. The percentages of the various minerals %
seaw
in the clay were estimated by comparing the intensity of Mineral by weight the c
characteristic peaks with the peak intensity of standard miner- tion
als. The results are presented in Table 1. Primary minerals incre
The mineralogy of the clay-size fraction was examined by Plagioclase orgar
running an X-ray analysis on samples made of the -2 Fm clay K-feldspar a facr
suspension allowed to air-dry slowly on glass slides for at least Qum the n
Amphibole Th
24 h. The X-ray diffraction pattern of an oriented air-dried Calcite
sample of the clay-size material is shown in Fig. 3b. As to be
expected, the proportion of phyllosilicate minerals is more Clay minerals the s;
Illite sulfa
important in the clay-size fraction. Rock flour minerals, how- Chlorite
ever, remain present in significantquantities. On the pattern of a tion i
Kaolinite quick
heat-treated sample (500C), the peaks at 7 A and 3.5A had
disappeared, thus confirming the presence of a small amount of (197!
kaolinite which becomes amorphous at 500C. The oriented and femc
method whereby the sulfate ion is precipitated with barium sitior
glycolated samples showed no trace of montmorillonite or other
chloride to form barium sulfate crystals. Chloride was measured meta:
smectites.
by potentiometry, i .e. by means of a chloride-sensingelectrode respc
Pore water chemistry coupled with a reference electrode and a pH-meter. The clays
The pore water used for the soluble salt determination was bicarbonate ion was determined by potentiometric alkalinity PrOcC
obtained by squeezing the soil in a modified consolidation titration with 0.01 N HCI. In all the pore water samples tested,
apparatus. The upper porous stone was covered with plastic and the concentration of C032- was small (less than 5% of the total Catic
the lower porous stone was replaced by a perforated plastic disc. alkalinity) and, therefore, its value was included in the Th
The water was collected from the base of the sample. A slow concentration of HC03-. All the above tests, with the exception of*
loading rate had to be used at the beginning of consolidation to of the chloride analysis, have an accuracy better than +5%. For of th,
avoid the squeezing out of liquefied soil between the piston and the potentiometric measurement of chloride, the accuracy is with
the consolidation ring. Overnight, about 6-10 g of water could about -C 10%. at p~
be obtained from the squeezing of some 80 g of clay at its initial The results of the pore water chemistry analyses are shown in amrn
water content. The ion concentration in the squeezed pore water Fig. 4. Sodium and bicarbonate are the predominant species in spect
was not found to be pressure dependent for the range of the pore water of La Baie clay. Their concentrations generally been
consolidation pressure used. increase with depth. In all the samples examined, the chloride Th
The concentrations of various ions were determined using concentration was less than 0.5 meq/L, with no particular trend meq/
standard procedures. Dissolved sodium, potassium, magnes- with depth. The concentrations of dissolved potassium, mag- of th
ium, and calcium were measured by atomic absorption spec- nesium, and calcium are also very low, showing a slight relati
trophotometry. Sulfate was determined by the turbidimetric increase with depth. conte
Whicl
 LESSARD AND MITCHELL 339

t FIG. 4. Pore water and adsorbed layer characteristics of La Baie clay measured 1 month after sampling.

The pore water chemistry analysis shows evidence of TABLE2. Comparison between the composition of seawater and the
I leaching and sulfate reduction, two conditions apparently composition of the pore water of the La Baie clay
necessary to the development of quickness. The composition of
the clay pore water is compared with the average composition of Seawater* Pore water Ratio
seawater in Table 2. Assuming a depositionalsalinity of 35 g/L, seawater/
the concentrations of all ions have decreased since the deposi- Ion meq/L % meq/L % pore water
tion of the clay, with the exception of bicarbonate, which
increased by a factor of 4 because of bacterial oxidation of Na' 459 77 11 95 42
organic matter. The chloride concentration has been reduced by K+ 10 1.7 0.3 2.6 33
a factor of 1800, which is indicative of the extensive leaching of
the material.
"f++
Ca
106
20
18
3.4
0.2
0.1
1.7
0.9
530
200
C1- 535 90 0.3 2.6 1800
The drastic depletion of calcium and magnesium is believed ~ 0 4 ~ - 55 9.3 2.0 17 27
to be due to precipitation as magnesium-calcite taking place in HC03- 2.3 0.4 9.3 80 0.25
the sulfate-reduction process, which also causes a decrease in
sulfate concentrations and an increase in bicarbonate concentra- *Values taken from Sverdrup et al. (1942).
tion and pH. The existence of sulfate reduction in a Champlain
quick clay was well demonstrated by Donovan and Lajoie higher specific surface areas, higher CEC, and higher water
(1979). In the absence of oxygen, dissolved sulfates, as well as content under a given effective stress.
femc oxides, are reduced bithe bacterially catalyzed decompo-
Exchangeable cations
uium sition of organic matter. This leads to the formation of
The exchangeable cation analysis is carried out by first
sured metastable iron monosulfide and ultimately pyrite, which is
removing the excess salt dissolved in the pore fluid by washing
trode responsible for the black mottles commonly found in quick
The with ethyl alcohol. The naturally adsorbed cations are then
clays. The chemical equilibria and reactions involved in the
linity process are described in more detail by Lessard (1981). exchanged for NH4+ by successive washing of the clay with a
:sted, solution of 1 N ammonium acetate at pH 7. The concentrations
Cation exchange capacity of the various cations introduced into the ammonium solution by
:total
1 the
The cation exchange capacity, CEC, expressed as meq/100 g replacement are then determined by atomic absorption spec-
,ption of dry soil, was determined by saturating the exchange complex trophotometry. Ammonium is used as the replacing ion,
of the clay with Na+, removing the excess Na+ by washing because the NH4+ ion is not commonly found in natural soils.
I . For
with ethyl alcohol, and replacing the adsorbed Na+ with NH4+ Five grams (dry weight) of wet soil was used for the anlaysis.
:cy is
at pH 7.0. The concentration of Na+ introduced into the One weakness of the method is that it may lead to an
wn in ammonium solution was then determined by atomic absorption overestimationof the adsorbed calcium as CaC03 is dissolved.
Spectrophotometry. Five grams (dry weight) of soil that had not In the case of the quick clay from La Baie, the pH of the soil was
ies in
been predried was used for the analysis. lowered from about 9.5 to a value of 7.0 by the addition of
;rally ammonium acetate. As the pH decreases, the solubility of
ioride The CEC measured for six samples varied from 9 to 13
meq/100 g, with an average of 11 meq/100 g, which is typical CaC03 is increased severalfold.
trend The test procedure used in the present analysis yielded values
of the Champlain Sea quick clay. As shown in Fig. 4, the
mag- relation of the CEC to depth follows the same trend as the water of about 30meq/100g for the "adsorbed" calcium. These
slight values are much higher than the CEC and are obviously not
content and the percentage of particles smaller than 2 Fm,
which is expected, since smaller particles are associated with representative of the actual amount of calcium present in the
340 CAN.G m H .I. VOL. 22. 1985
- - -,
. --
6.0 76
double layer of the natural clay. The distributions of adsorbed I

sodium, potassium, and magnesium with depth follow the same -- .*--_
A SO .. - 3 1- -1.+ -----~-
- - - 8--
trend as the variation of the CEC (Fig. 4). Sodium makes up % 2.0 '
0 - =
.' - -
0
0

&;1.0 - -
nearly 50% of the exchangeable cations, and potassium about
.,
--A-
Lo'
,' 0
16%. The concentration of adsorbed magnesium, which makes i q ,
I I
o I I

up about 30% of the total exchangeable cations, is very high 0.5 6 1,"" ;
'
compared with the amount of magnesium in the free pore water. g 1 2 at,'
Y
2.0-

Although the double-layer theory predicts a larger proportion of 20 o.2 - X

<
LO'
"
I l . ~-
divalent cations in the adsorbed system than in the equilibrium 2 :
.; e
solution, it is probable that at least a portion of the measured ' .1:; : < O 0
-
adsorbed magnesium came from the dissolution of magnesium- 1 I I = F 6.0- .-
0

calcite. $ 2 - -D B

Storage procedures
7

(I
1
1 I $
a
E
2.5-

o
1
-
,:~o'(--
,,,6
0--------~-

UNWAXED SAMPLES:
On
Y
..
Upon reception, the samples, which had been waxed with - I I
I l
0 AIR
- 0 5.0-
a'
0 OXYGEN
several layers of paraffin, were stored in a wet room at 20C and ,Y '
f 1
I A NITROGEN
z
nd ,,
100% relative humidity. This represents the storage procedure ?- I 1
@ - --- - --.0
8 -
most frequently used in soil laboratories. For that reason it will - 1 1 A - - 0 n - - - 8 - - - - - - -I- - A
&--+
I
, '
g 0
be referred to as "standard" procedure. Two months after I
i
' ??
sampling, several samples were removed from the wet room and '2- 1 '\

stored under "special" storage conditions. 1

I
10- s----~.. '.
'-.I:--- ----- -
In order to better understand the role of oxidation in the aging - - -----.!I
phenomenon, several samples were unwaxed and stored at 20C o I I 0 - I

-----a-
in airtight plastic containers. This type of storage might be used lo,o -0----

for sample trimmings or with material after other tests have been ,
,
? a - - - --- 8--
>----
performed. These samples are often used for Atterberg limit
,,&-
tests. A total of twelve 400 g samples were placed in three sealed
I,$,:
containers filled with air, nitrogen, and oxygen. The nitrogen 5
o.o
1 -. -0--
__--- --A-

and oxygen atmospheres were obtained by flushing the con- 0.5 wl;-o" - - - - - - ---a'-
tainer with the appropriate gas for about 5 min. The degree of ,
, $
o 25 - 1,:-
e
,p
oxygen depletion was not assessed. A bottle of water was placed
o I 1
in each of the containers to maintain 100% relative humidity. 7.5
0 5 10 15 0 5 10 15
In order to study the effect of temperature on aging, six TIME AFTER SAMPLING -MONTHS TlME AFTER SAMPUNG - MONTHS

200g were at 40C. After months of FIG.5 . Effectof time on the properties of samples unwaxed 2 months FIG
the wax on two of the samples was removed, in order to samplingand stored in air, 02,and N~ at 2 0 0 ~ .
dter 20C.
accelerate the rate of aninn. circle
In an attempt to measurethe influence of bacterial activity on brownish. The samples had fissured and seemed stronger and
a ing, four 400 g samples were sterilized using X-rays from a stiffer as if they had dried, but measurement of the water content and
C
' o source. The waxed undisturbed samples were irradiated I acco.
showed no significant change during the storage period. The
for 24 h, receiving a total dose of 0.2 MGy (20 ~ r a d ) . A
' dose undisturbed strength increased by 10-20% during aging. seal.
of 0.02-0.04 MGy (2-4 Mrad) is generally recommended for However, this might be due to the loss of sensitivity, which as dil
100% sterilization (McLaren et al. 1957, 1962; McLaren increases the adhesion along the fall-cone. aPPa
1969). The effect of irradiation on the geotechnical and The remoulded strength increased markedly with time, from
chemical properties of the clay is minimal. A visual examina- whereas the liquidity index and the pH decreased, as shown in ofth
tion of one sample immediately after sterilization showed the Fig. 5. The concentration of all ions in the pore water increased The
same extreme quickness as before. The sterilized samples were also
with time, with the exception of bicarbonate, which shows a
stored in standard conditions, i.e., waxed, at 20C, and 100% downward trend beginning some 3 months after removal of the easil
relative humidity. wax coating. The chloride concentration was found to remain X-ra
The changes in properties of all the samples stored under Undc
very low during aging and for that reason has not been plotted. It
"standard or "special" conditions were measured by periodic is noteworthy that the use of nitrogen did not slow down the whic
tests carried out over a 12 month period. At the end of the testing aging process. It appears that the amount of residual air in the the k
program on aging, one sample that had been left in the original container or possibly seeping through the "airtight" seal is to th
Shelby tube for a period of 14 months was also tested. quite sufficient to achieve considerable oxidation of the sam- wax
pies. In this regard, it should be pointed out that the greatest a ral
Results Sam1
scattering in the results was obtained between two samples from
Unwaxed samples about the same depth, stored 114 months in the same N,-filled a ye;
After 3 months' storage in airtight plastic containers, in the plastic container. In
presence of air, oxygen, and nitrogen, all samples were partially chen
covered with whiteand orange mold and with brown to black Waxed samples by tt
spores. The clay, originally bluish gray in colour, had turned As shown in Fig. 6, the changes in geotechnical properties chan
and pore water chemistry of waxed samples stored at 20C were col01
he gray (Gy) is the absorbed dose that is equal to one joule per less pronounced than in unwaxed clay (Fig. 5). The wax slowed none
kilogram (J/kg). down the aging process but did not prevent it. With time, more for t
 LESSARD PLNDMITCHELL 34 1
soluble iron and dissolved organic carbon (DOC)was made on
1 'STANDARD' SAMPLES
4 10 pore fluid samples showing various degrees of colour
changes. In all the samples, the concentration of dissolved iron
was below measurable value, which is in agreement with the
low solubility of iron at the pH of the pore water (pH = 8.3).
The dissolved organic carbon analysis of the pore water samples
yielded values ranging between 17 and 95 ppm, higher values
being measured in the yellowish fluids. No further analysis was
carried out to identify the exact nature of the organic compound.
However, there is a strong possibility that the yellow substance
consists of fulvic or humic acid formed by the oxidation of
organic matter as complex compounds are broken down into
organic acids. It is likely that these organic acids are the source
of the yellow stains appearing in the wax during aging.
Sterilized clay
The values of the physicochemical properties of sterilized
samples fall, with a few exceptions, in the range of variation
obtained for the samples waxed and stored at 20C as shown in
Fig. 6. The samples tested 8 and 14 months after sampling
exhibited higher alkalinity and lower pH, as compared with the
average "standard" sample, apparently caused by oxidation of
organic matter as indicated by the presence of yellow stains on
the wax and the marked yellowish colour of the pore water.
Although the dosage used was about five times that recom-
mended for 100% sterilization, the irradiation of quick clay
samples did not prevent aging from proceeding. It is therefore
likely that aging is not directly mediated by bacteria. However,
enzymes, which are smaller than bacteria and less sensitive to
radiation (McLaren and Reshetko 1957), may act as catalysts in
the oxidation process. Molds and spores are also more resistant
to sterilization and might remain active after irradiation.
0 5 10 15 0 5 10 15
TIME AFTER SAMPLING - MONTHS TIME AFTER SAMPLING - MONTHS
Efect of temperature
FIG.6. Effect of time on the properties of waxed samples stored at The effect of temperature on aging of waxed and unwaxed
20C. The brackets indicate the range of variation of the results and the samples is illustrated in Fig. 7. The physicochemical properties
circles represent the average value of each series of tests. of the waxed samples stored at 4C showed very little variation
r and with time over a period of 1 year. No yellow stain was observed
and more samples lost their quickness. This was always on the wax and the pore water remained colourless. The samples
mtent
accompanied by changes in the colour and condition of the wax exposed to the air at 4C showed clear signs of aging. They
. The
seal. Yellowish stains appeared in the originally white paraffin
ging. as diffuse spots randomly distributed over the sample or as lines,
looked drier and stiffer but with no apparent loss of water
vhich content. Contrary to those stored at 20C,however, no molds or
apparently along microfissures in the wax. These may originate spores were found on the samples. The pH also remained
from the attack of wax by organic acids formed as by-products relatively high, and the bicarbonate showed no increase in the
time,
of the oxidation of complex carbohydrates present in the clay. early stage of aging as observed at 20C.This suggests that low
wn in
The formation of a black coating at the soil-wax interface was temperature hinders the oxidation of organic matter. However,
2ased
also observed, mostly at the ends of the samples where air is the increase of sulfate with time of exposure to air does not seem
)WS a
I easily trapped between the wax and clay surface. This black to be affected by storage temperature, as illustrated by Fig. 8.
,f the
X-ray amorphous compound is believed to be of organic origin. The figure also shows that in the first few months of aging the
.main Under optical microscope it appeared to be formed of spores,
ed. It concentration of calcium is lower at 4OC,apparently because of
which are known to grow in aerobic environments. In summary, the limited organic matter oxidation.
n the
the large scattering of the results shown in Fig. 6 seems to be due
n the Sample kept in Shelby tube
:al is to the variable quality of the wax coating. Microfissures in the
wax or air pockets trapped at the clay-wax interface resulted in The last sample analyzed in the testing program on aging was
sam- the one shipped and stored in the Shelby tube. The sample had
a rapid oxidation of some samples, while some well-sealed
:atest been prepared as follows: Immediately after sampling, 2-3 cm
samples showed little sign of aging even after storage for over
from of clay was removed from the lower end of the tube and the tube
filled a year.
In all the samples tested, the changes in the geotechnical and was sealed by means of a plastic and rubber "Soilseal" covered
chemical properties of the material were accompanied not only with about 1cm of wax. The other end was sealed by pouring
by the presence of yellow or black stains in the wax but also by a 2-3 cm of wax in the tube. Upon reception, the sample was
trties change in the colour of the squeezed pore water. Initially stored in a wet room at 20C.
were colourless, it became gradually yellow as aging occurred. Since Fourteen months after sampling, the Shelby tube was opened
wed none of the measured ion concentration increases could account and the clay tested. Surprisingly, the sample was found to be
more for the change in colour of the pore fluid, a determination of still extremely quick. The very high pH and low concentration
 342 CAN. GEOTEXR.J. VOL. 22, 1985

5.0

SAMPLE
WAXED
UNWAXED
.
QC

o
20c

0
I
,/'-
-

I
0
W '0

TIME OF EXPOSURE T O AIR - MONTHS

FIG.8. Effect of time on the pore water concentrationsof sulfate and

calcium in samples exposed to air (Po,= 20 kPa (0.2 atm)).

7 51 I I I I I 1 [41
5 10 15 0 5 10 15
TlME AFTER SAMPLING - MONTHS TlME AFTER SAMPLING - MONTHS The
cata
FIG. 7. Effect of time on the properties of waxed and unwaxed oxia
samples stored in air, at 4OC, and 20C. The arrows indicate when the
wax was removed from the samples. quic
, the
of dissolved ions, particularly S042-, Ca2+, and Mg2+, matt
indicated that virtually no aging had occurred in more than 1 T
year. A visual examination of the sample also showed no sign of prec
alteration. The clay was medium gray, slightly bluish in colour; ' Dl
no brownish ring of oxidation was found on the periphery of the STORAGE CONDITIONS

sample. The steel itself did not show any sign of oxidation inside
the tube. Apparently, the thick wax caps are less susceptible to
the formation of small fissures (possibly caused by handling or
SAMPLE

WAXED
UNWAXED 0
.
4'C 2@C

0
* z o o

1
or b
[61
temperature change during transport and storage) than is the 0

standard multilayer wax coating and thus provide a much better G

barrier against oxygen diffusion. amc
I invc
Chemical processes in aging FIG.9. Effect of soil pH upon the concentrationsof bicarbonate and The
The evolution of sulfate and bicarbonate concentrations with sulfate in the pore water of samples stored under various conditions. I of t1
time in waxed and unwaxed samples suggests that two major SUP1
oxidation processes take place in the clay during aging, namely Fe((
the oxidation of organic matter and the oxidation of iron sulfide. In the waxed samples, CO2 cannot escape and builds up. The righ
increase in partial pressure of C02 results in the formation of a oxic
Organic matter oxidation weak acid, H2C03, which decreases the pH and increases the the
As illustrated by Fig. 9, the bicarbonate concentration, for a concentration of calcium and bicarbonate by dissolving calcium diag
given pH, is much higher in the waxed samples (solid symbols) carbonate according to the reaction , of s
than in the unwaxed samples (open symbols). This is due to the becc
fact that the oxidation of carbohydrates or organic matter , pres
(generalized here as a simple formaldehyde CH20) produces The oxidation of organic mattter appears to be catalyzed by
C 0 2 according to the general reaction Chc
microorganisms such as spores or molds that show reduced T
activity at lower temperature. This is in agreement with the obt2
I LESSARD AND MITCHELL
I
FIG. 10. Mechanism of pyrite oxidation.
absence of organic growth and the low bicarbonate production
observed on the samples stored at 4OC.
100
Iron sulfide oxidation
In the unwaxed samples the C02produced by the oxidation of
Ifate and carbohydrates is free to diffuse out of the sample and therefore
has little influence on the pore water chemistry of the clay. The
7
bicarbonate concentration remains low (Fig. 9). In these
samples where oxygen is readily available, the drastic increase
in sulfate concentration is believed to result from the oxidation
of iron sulfide through a chain reaction as is illustrated by Fig.
10 which shows the various steps involved in the cyclical
oxidation of pyrite. First, pyritic sulfur is directly oxidized by
oxygen according to the equation
Ferrous iron is then oxidized to femc iron by the reaction
.o
I
The reaction in [4] occurs extremely slowly at low pH unless it is
catalyzed by bacterial activity involving Thiobacillus ferro-
I
oxidans (Stumm and Morgan 1970; Zajic 1969). In the case of
I quick clays, the rate of reaction may be significant, even in
the absence of bacterial activity because of the high pH of the
material (Snoeyink and Jenkins 1980).
The femc iron formed by the reaction in [4] may either
precipitate as Fe(OH)3 according to the reaction
[5] 4Fe3+ + 12H20+ 4Fe(OH),,,) +12H+
or be available to oxidize more FeS2 to ~ e by~the+reaction
Once the pyrite oxidation has started, i.e. once a small
amount of FeS2 has been oxidized by O2 ([3]), oxygen is
0
involved only indirectly in the reoxidation of Fe2+ to Fe3+ (141).
The oxidation of FeS2(,) ([3]) is no longer of significance, most
te and
ions. of the FeS2 being oxidized by Fe3+ according to [6]. An ample
supply of Fe3+ for this reaction is guaranteed, because
Fe(OH)3(,, precipitated by the reaction in [5] will be present
right at the site of iron oxidation. The net result of pyrite
. The
iof a oxidation is the formation of yellow Fe(OH)3(s),which causes
the brownish discoloration of the samples. Based on a Eh-pH
s the
cium diagram representing the stability fields of iron in the presence
of sulfur and carbonate, the level of oxygen for which FeS2
becomes unstable relative to Fe(OH), corresponds to a partial
pressure of oxygen in the order of kPa.
d by Changes in pore water chemistry
uced The global equation describing the oxidation of pyrite may be
I the obtained by summing up [3]-[5], which yields
343
Similarly the oxidation of iron monosulfide follows the reaction
In neutral or slightly acid waters, Fe(OH)3 is essentially
insoluble, and therefore the major changes in the pore water
chemistry will be brought about by the formation of sulfuric
acid, which causes a decrease in pH and alkalinity as the
bicarbonate is neutralized by the strong acid according to the
reaction
However, the system is buffered by the presence of calcite,
which is gradually dissolved by the sulfuric acid following the
equation
The iron sulfide oxidation does not appear to be catalyzed by
microorganisms, since neither sterilization nor low storage
temperature had any effect on the rate of sulfate formation.
The scattering of the results observed in Fig. 9 is mostly due
to the effect of the various storage procedures upon the
oxidation of organic matter and iron sulfide. If the contribution
of the two processes is taken into account by plotting the
summation of S042- and HC03- (which is approximately equal
to the summation of cations given the low chloride concentra-
tions) as a function of pH, a much better correlation is obtained
(Fig. 11). Again, however, the higher concentrations are
obtained in the unwaxed samples apparently because sulfuric
acid predominantly produced in these samples is a strong acid
and dissolves more CaC03 than does H2CO3. For similar
reasons, the pore water concentration of calcium is also slightly
higher in unwaxed samples at a given pH.
Cation exchange
Although the increase in the total salinity of the clay is due to
the dissolution of CaC03 by carbonic and sulfuric acids, only a
fraction of the cations found in the pore water is calcium, as
shown by inference in Fig. 11. This is due to the cation
exchange process which occurs in the clay. At the beginning of
aging, i.e. at pH = 9.5, the calcium makes up only 0.5% of the
pore water cations. As aging goes on, most of the calcium
produced by dissolution of calcite is adsorbed by the double
layer. Gradually, an equilibrium is reached and more calcium
remains in the "free" pore water. At a pH lower than 8.5,
calcium makes up 5-10% of the dissolved cations.
Figure 12 shows how the dissolution of calcium and its
increasing pore fluid concentration affects the concentration of
sodium, potassium, and magnesium in the pore water. In the
early stage of aging, the monovalent cations are preferentially
displaced from the adsorbed system as predicted by the
double-layer theory. The relationship between the calcium and
magnesium concentrations is more linear because they have the
same valence.
As shown in Fig. 13, the increase in salinity and particularly
in the pore water concentration of divalent cations results in a
decrease of the interparticle forces and leads to an increase in
remoulded strength (Mitchell 1976).
Effect of amorphous material
Although the changes in geotechnical properties observed in
the quick clay during aging can be adequately explained by the
 344 CAN. GEOTECH. 1. VOL. 22, 1985

- - ec
SAMPLE 2o.c -
STORAGE CONDITIONS
Ot -
.
WAXED 8
UNWAXED 0 0

O 0 WAXED
UNWAXED D 0

FIG. 13. Relationship between remoulded strength and the con-

centrations of divalent cations and divalent plus monovalent cations in
0.2 - samples stored under various conditions.

7.5 8.0 8.5 9.0 9.5 10.0

OH

70 meq/L (the maximum concentration measured in 1 year of

STORAGE CONDITIONS
FIG. 12. Effect of dissolved calcium concentration upon the
of sodium, potassium, and magnesium in the pore water
of samples stored under various conditions.
aging), 0.05-0.1% of Fe per dry weight of soil will be
precipitated as Fe(OH), by the oxidation of iron sulfide. Such an
increase in Fe might lead to a slight increase in remoulded
strength and liquid limit by increasing the specific surface area
and the cation exchange capacity (Hendershot and Carson
1978). The contribution of Fe(OH)3 to remoulded strength is
likely to increase as aging proceeds.
It is very difficult to estimate the amount of Fe(OH)3 present
in the clay other than by calculation. The selective dissolving
technique (Yong et al. 1979) often used to estimate the amount
of amorphous material is not believed to be adequate to detect
such a small increase in Fe(OH), in the clay. In effect, the
method, which is based on alternate washing of the clay with
0.5 N NaOH and 8 NHCI, results in an overestimation of the
amount of Fe(OH)3 actually present in the soil, as the strong
acid causes the oxidation of FeS and possibly of a large part of
FeS2. Crystalline Fe203 and Fe30, present in relatively low
quantities may also contribute to an overestimation of the
amorphous compounds. Whether these crystalline oxides may
act as cementing compounds is not known. As pointed out by
Quigley (1980), the selective dissolution technique also results
in the dissolution of chlorite and illite, which is likely to cause
an overestimation of amorphous alumina and silica unless
appropriate are made. These facts should be kept in
mind when examining the relationship between sensitivity and
the so-called amorphous compounds.
 LESSARD AND MITCHELL
r.
.I Conclusion and recommendations
( m e physicochemical properties of a clay from La Baie were
and to be typical of Champlain quick clays: very low
lasticity, liquidity index greater than 3, sensitivity greater than
ICATIONS
80, and pH of about 9.5. The mineralogy is characterized by
abundance of primary or "rock flour" minerals such as
ddspar, ~ quartz, amphibole, and calcite, and by illite as the
-.
3
AMPLE
+AI(ED
~ COND~TIONS
.
ec ,
~ ~ a ~
, cincipal clay mineral. The dissolved pore water chemistry
~ n s i s predominantly
t~ of sodium bicarbonate. Low concentra-
INWAXED O
ens of sulfate, calcium, and magnesium and high alkalinity
pd pH are attributed, in part, to the anaerobic sulfate-reduction
u , , , , , ,>mess taking place in situ. This process is believed to be
I W 200
" bntrolled by the bacterially catalyzed decomposition of organic
aatter and results in the formation of metastable iron sulfide.
'ngth and the cQ Some signs of aging were observed in all the quick clay
'novalentcationsbples stored in the laboratory, regardless of the method of
torage. With time, the remoulded strength and liquid limit
ocreasedand the sensitivity, liquidity index, and pH decreased.
'he water content, plastic limit, and undisturbed strength
uble comPounkmained practically unchanged. The pore water concentrations
xistrength- Thlf calcium, magnesium, and sulfate increased severalfold.
!during aging b From the test results it is postulated that the aging of quick
referred hays occurs as follows: In the presence of oxygen, the organic
.cqstallinit~,hatter is oxidized to form carbonic acid, which dissolves
the 9 whiwcium carbonate, thus increasing the concentration of calcium
(Loken 197bd bicarbonate in the pore water. In the meantime, pyrite and
capable kr)iron monosulfide are oxidized to sulfuric acid and yellow
'"8 (Yo% et alfon hydroxide, which imparts a brownish colour to the oxidized
:lay. The sulfuric acid produced by the cyclical oxidation of
:aging may blfon sulfide neutralizes the bicarbonate and dissolves CaC03. A
s m d l n t h e p o k g e proportion of the dissolved calcium is adsorbed in the
'Ifate and I mdouble layer, thereby displacing sodium, potassium, and
losulfide ~ieldtna~nesium from the adsorbed system to the free pore water.
hatforasamplne increase in salinity and especially in divalent cations
'ncentration &sulting from the dissolution of CaC03 reduces the inter-
in
'ed 1 Year +article repulsive forces and increases the remoulded strength.
will None of the storage procedures studied was found to be
ulfide. Such ntirely satisfactory. Any sample stored in the laboratory will
in remould~ventuallyundergo significant changes in pore water chemistry
ic surface areaPd in geotechnical properties. Consequently, it is recom-
a d Carsonmended that the clay be tested as soon as possible after it is
led ifobtained from the field.
If it is necessary to store quick clay samples for a period in
(OH)! ~resen(nrcessof a few months, certain measures may be taken in order
Ive dlssolvmg to minimize the rate of aging. The following storage procedures
Ite the ~ ~ o uare n lsuggested:
Pateto detect (1) The samples should be stored at 4C to reduce the organic
In effect* he'matter oxidation to a minimum and prevent rapid deterioration
clay wihlof the wax coating.
mation of the (2) Quick clay samples in Shelby tubes may be stored in the
as the strong tube if the steel is in good condition (no rust). The tube should be
' large Part of sealed with thick wax caps (2-3 cm) at both ends to reduce the
elativel~low oxygen diffusion.
lation of the (3) Samples extruded from the Shelby tube should be waxed
2.0xides may with many layers of wax. Special care should be taken not to
out by trap air pockets between the clay and the wax.
results (4) Block samples should be precut to the size needed for the
k e l to ~ cause test to be carried out in the laboratory, and then waxed
unless individually, in order to avoid exposing the entire block to air
'Id be kept in every time a test is to be performed.
nsitivit~a d It is recommended to measure the pH of the clay at the time of
sampling or at the reception of the samples, in order to monitor
345
the aging of clays stored in the laboratory. In effect, the pH is
very sensitive to oxidation in the early stages of aging.
Furthermore, its measurement was found to be simple and
reproducible if the same method is used for all tests.
For determining the pore water chemistry of marine clays, not
only should the extraction of the pore water be done soon after
sampling but the chemical analysis should also be carried out as
soon as possible after squeezing the water out of the sample, in
order to minimize potential contamination by the aging of solid
particles in suspension in the squeezed pore water. If storage
becomes necessary, the water extract can be frozen to slow
down the rate of oxidation of organic matter and iron sulfide.
Acknowledgments
The work described in this paper would not have been
possible without the support of many individuals and organiza-
tions. Financing was provided by the Natural Sciences and
Eqgineering Research Council of Canada and the Minist2re de
1'Education du QuCbec. The samplgs were provided through the
courtesy of the Minist2re de 1'Energie et des Ressources,
Gouvemement du QuCbec. Jacques Lebuis, director of the
"Service de la Gt5otechnique" was responsible for the sampling
program and provided the writers with helpful information on
quick clays from Quebec. His collaboration in the research
project is gratefully acknowledged.
BJERRUM, L., and ROSENQVIST, I. TH. 1956.Some Experiments with
artificially sedimented clays. Geotechnique, 6, pp. 124- 136.
DONOVAN, J. J., and LAJOIE, G. 1979.Geotechnical implications of
diagnetic iron sulfide formation in Champlain Sea sediments.
Canadian Journal of Earth Science, 16, pp. 575-584.
FOSCAL-MELLA, G. 1976. Analyse min6alogique des vgiles gla-
ciaires. Th5se M.Sc.A., Dtpartment de genie mintral, Ecole Poly-
technique, Montn5al, Qu6.
HENDERSHOT, W. H., and CARSON, M. A. 1978. Changes in the
plasticity of a sample of Champlain clay after selective chemical
dissolution to remove amorphous material. Canadian Geotechnical
Journal, 15, pp. 609-616.
HESSE,P. R. 1971. A textbook of soil chemical analysis. Chemical
Publishing Company Inc., New York, NY, pp. 52-53.
LESSARD, G. 1981. Biogeochemical phenomena in quick clays and
their effects on engineering properties. Ph.D. dissertation, Depart-
ment of Civil Engineering, University of California, Berkeley, CA,
335 p.
LOKEN,T. 1970. Recent research at the Norwegian Geotechnical
Institute concerning the influence of chemical additions on quick
clay. Geologiska Foereningens i Stockholm Foerhandlingar, 92,
pp: 133-147.
MCKYES,E., SETHI,A. J., and YONG,R. N. 1974. Amorphous
coatings on particles of sensitive clay soils. Clay and Clay Minerals,
22, pG 427-433.
MCLAREN. A. D. 1969.Radiation as a technique in soil biology and
biochemistry. Soil Biology & Biochemistry, 1, pp. 63-73.
MCLAREN, A. D., and RESHETKO, L. 1957. Soil sterilized by
irradiation. California, Department of Agriculture, Bulletin, 11,
p. 12.
MCLAREN, A. D., RESHETKO, L., and HUBER, W. 1957.Sterilization
of soil by irradiation with an electron beam, and some observationon
soil enzyme activity. Soil Science, 83, pp. 497-502.
MCLAREN, A. D.,LUSE,R. A., and SKUJINS, J. J. 1962.Sterilization
of soil by irradiation and some further observation on soil enzyme
activitv. Soil Science Society of America, Proceedings, 26, pp.
371-i77.
MITCHELL, J. K. 1976.Fundamentals of soil behavior. John Wiley &
Sons, Inc., New York, NY, 422 p.
346 CAN. GEOTECH. J. VOL. 22, 1985

QUIGLEY, R. M. 1980. Geology, mineralogy, and geochemistry of TAVENAS, F., CHAGNON, J.-Y., and LA ROCHELLE, P. 1971. The
Canadian soft soils: a geotechnical perspective. Canadian Geotech- Saint-Jean-Vianney landslide: observations and eyewitnesses ac-
nical Journal, 17, pp. 261-285. counts. Canadian Geotechnical Journal, 8, pp. 463-478.
RANKIN,W. L. 1970. Suggested method of test for organic matter TORRANCE, J. K. 1976. Pore water extraction and the effect of sample
content of soils by redox titration. American Society for Testing and storage on the pore water chemistry of Leda clay. Soil Specimen
Materials, Special Technical Publication, No. 479, pp. 286-287. Preparation for Laboratory Testing, American Society for Testing
SNOEYINK, V. L., and JENKINS,D. 1980. Water chemistry. John and Materials, Special Technical Publication, No. 599, pp. 147-
Wiley & Sons, New York, NY, 463 p. 157.
S~DERBLOM, R. 1969. Salt in Swedish clays and its importance for YONG,R. N., SETHI,A. J., and LAROCHELLE, P. 1979. Significance
quick-clay formation. Results from some field and laboratory of amorphous material relative to sensitivity in some Charnplain
studies. Swedish Geotechnical Institute, Proceedings, No. 22,63 p. clays. Canadian Geotechnical Journal, 16, pp. 5 11-520.
1974. Aspects on some problems of geotechnical chemistry, YONG,R. N., SETHI,A. J., and SUZUKI,A. 1980. Contribution of
Part 111. Swedish Geotechnical Institute, Proceedings, No. 55, pp. amorphous material to properties of a laboratory-prepared soil.
452-468. Canadian Geotechnical Journal, 17, pp. 440-446.
STUMM,W., and MORGAN, J. J. 1970. Aquatic chemistry, Wiley ZAJIC,J. E. 1969. Microbial biogeochemistry. Academic Press, New
Interscience, New York, NY. York, NY, 345 p.
SVERDRUP, H. U.,JOHNSON, M.W., and FLEMING, R. H.1942. The
oceans, their physics, chemistry and general biology, Prentice-Hall
Inc., New York, NY.

rep
I fac
me
I
dip
sta
ass
ah
, nor
strc
, exi
inc
, tra
It
, WC

1 tw
ste
1
aP
ral
I dil

I
se
ISC